Cancer is often the outcome of interaction between external exposures or lifestyle factors, and genetic susceptibilities. These processes can be studied through epidemiology, and through modeling in animals. Nasopharyngeal cancer is common only in certain geographically-delimited populations, with lifestyle factors, especially diet, implicated in causation, and genetics and infections as secondary factors. Causative chemicals in the diet may include N-nitrosamines, especially N- nitrosodimethylamine, which is activated in cells by cytochrome P450 2E1 (CYP2E1). We have found a striking correlation between incidence of nasopharyngeal cancer in Chinese and an allelic form of the CYP2E1 gene, c2c2, that may be highly expressed. However, this association was limited to nonsmokers. Among smokers, incidence of the disease was less, and was positively associated with the other allelic form of CYP2E1, c1c1. Thus a complex interplay of several exposure factors with genetic constitution determined cancer outcome. To begin to understand the role of the CYP2E1 gene in more detail, we have examined part of the structure of this gene in two species of monkey, patas and cynomolgus. The polymorphic promoter region of the gene in the patas monkeys is very similar to that of the c2c2 humans, whereas that of the cynomolgus monkeys, while still highly homologous, is much longer and has a different base sequence in the diagnostic RFLP site. The patas and human both have evolved on the savannah of Africa, whereas the cynomolgus is an arboreal species. The patas monkey has potential as a model for the susceptible c2c2 humans. We have used modeling in mice to begin to approach the question of promotion of lung tumors, suspected to occur during the long latency of human lung cancers, and a possible target for intervention strategies. Since the K-ras gene is often mutated in both murine and human lung adenocarcinomas, we have started to examine the levels and cell membrane localization of K-ras p21 in mouse lung as a function of genetics and of exposure to the tumor promoters, 2,3,7,8,-tetrachloro-dibenzo-p- dioxin (TCDD) and polychlorinated biphenyls. We discovered, for the first time, marked differences among various genetic mouse strains. Whereas the total K-ras p21 was similar in all strains, amounts in membrane were significantly greater in C57BL/6 and BALB/c mice, compared with other strains. Treatment with TCDD caused a significant increase in the p21 in lung cell membranes in some but not all mouse strains. The relevance of these changes in genetic susceptibility to tumor promotion is under study.